-
Notifications
You must be signed in to change notification settings - Fork 81
/
Copy patharray_api.html
481 lines (402 loc) · 31.8 KB
/
array_api.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
<!DOCTYPE html>
<!-- data-theme below is forced to be "light" but should be changed if we use pydata-theme-sphinx in the future -->
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" data-content_root="../" data-theme="light"> <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" data-content_root="../" data-theme="light"> <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1" />
<meta property="og:title" content="11.1. Array API support (experimental)" />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://scikit-learn/stable/modules/array_api.html" />
<meta property="og:site_name" content="scikit-learn" />
<meta property="og:description" content="The Array API specification defines a standard API for all array manipulation libraries with a NumPy-like API. Scikit-learn’s Array API support requires array-api-compat to be installed. Some sciki..." />
<meta property="og:image" content="https://scikit-learn.org/stable/_static/scikit-learn-logo-small.png" />
<meta property="og:image:alt" content="scikit-learn" />
<meta name="description" content="The Array API specification defines a standard API for all array manipulation libraries with a NumPy-like API. Scikit-learn’s Array API support requires array-api-compat to be installed. Some sciki..." />
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>11.1. Array API support (experimental) — scikit-learn 1.5.dev0 documentation</title>
<link rel="canonical" href="http://scikit-learn.org/stable/modules/array_api.html" />
<link rel="shortcut icon" href="../_static/favicon.ico"/>
<link rel="stylesheet" href="../_static/css/vendor/bootstrap.min.css" type="text/css" />
<link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
<link rel="stylesheet" href="../_static/copybutton.css" type="text/css" />
<link rel="stylesheet" href="../_static/plot_directive.css" type="text/css" />
<link rel="stylesheet" href="../https://fonts.googleapis.com/css?family=Vibur" type="text/css" />
<link rel="stylesheet" href="../_static/jupyterlite_sphinx.css" type="text/css" />
<link rel="stylesheet" href="../_static/sg_gallery.css" type="text/css" />
<link rel="stylesheet" href="../_static/sg_gallery-binder.css" type="text/css" />
<link rel="stylesheet" href="../_static/sg_gallery-dataframe.css" type="text/css" />
<link rel="stylesheet" href="../_static/sg_gallery-rendered-html.css" type="text/css" />
<link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
<script id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
<script src="../_static/js/vendor/jquery-3.6.3.slim.min.js"></script>
<script src="../_static/js/details-permalink.js"></script>
</head>
<body>
<nav id="navbar" class="sk-docs-navbar navbar navbar-expand-md navbar-light bg-light py-0">
<div class="container-fluid sk-docs-container px-0">
<a class="navbar-brand py-0" href="../index.html">
<img
class="sk-brand-img"
src="../_static/scikit-learn-logo-small.png"
alt="logo"/>
</a>
<button
id="sk-navbar-toggler"
class="navbar-toggler"
type="button"
data-toggle="collapse"
data-target="#navbarSupportedContent"
aria-controls="navbarSupportedContent"
aria-expanded="false"
aria-label="Toggle navigation"
>
<span class="navbar-toggler-icon"></span>
</button>
<div class="sk-navbar-collapse collapse navbar-collapse" id="navbarSupportedContent">
<ul class="navbar-nav mr-auto">
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../install.html">Install</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../user_guide.html">User Guide</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" href="classes.html">API</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" href="../auto_examples/index.html">Examples</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link" target="_blank" rel="noopener noreferrer" href="https://blog.scikit-learn.org/">Community</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../getting_started.html" >Getting Started</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../tutorial/index.html" >Tutorial</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../whats_new/v1.5.html" >What's new</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../glossary.html" >Glossary</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../developers/index.html" >Development</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../faq.html" >FAQ</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../support.html" >Support</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../related_projects.html" >Related packages</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../roadmap.html" >Roadmap</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../governance.html" >Governance</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="../about.html" >About us</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="https://github.com/scikit-learn/scikit-learn" >GitHub</a>
</li>
<li class="nav-item">
<a class="sk-nav-link nav-link nav-more-item-mobile-items" href="https://scikit-learn.org/dev/versions.html" >Other Versions and Download</a>
</li>
<li class="nav-item dropdown nav-more-item-dropdown">
<a class="sk-nav-link nav-link dropdown-toggle" href="#" id="navbarDropdown" role="button" data-toggle="dropdown" aria-haspopup="true" aria-expanded="false">More</a>
<div class="dropdown-menu" aria-labelledby="navbarDropdown">
<a class="sk-nav-dropdown-item dropdown-item" href="../getting_started.html" >Getting Started</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../tutorial/index.html" >Tutorial</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../whats_new/v1.5.html" >What's new</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../glossary.html" >Glossary</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../developers/index.html" >Development</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../faq.html" >FAQ</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../support.html" >Support</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../related_projects.html" >Related packages</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../roadmap.html" >Roadmap</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../governance.html" >Governance</a>
<a class="sk-nav-dropdown-item dropdown-item" href="../about.html" >About us</a>
<a class="sk-nav-dropdown-item dropdown-item" href="https://github.com/scikit-learn/scikit-learn" >GitHub</a>
<a class="sk-nav-dropdown-item dropdown-item" href="https://scikit-learn.org/dev/versions.html" >Other Versions and Download</a>
</div>
</li>
</ul>
<div id="searchbox" role="search">
<div class="searchformwrapper">
<form class="search" action="../search.html" method="get">
<input class="sk-search-text-input" type="text" name="q" aria-labelledby="searchlabel" />
<input class="sk-search-text-btn" type="submit" value="Go" />
</form>
</div>
</div>
</div>
</div>
</nav>
<div class="d-flex" id="sk-doc-wrapper">
<input type="checkbox" name="sk-toggle-checkbox" id="sk-toggle-checkbox">
<label id="sk-sidemenu-toggle" class="sk-btn-toggle-toc btn sk-btn-primary" for="sk-toggle-checkbox">Toggle Menu</label>
<div id="sk-sidebar-wrapper" class="border-right">
<div class="sk-sidebar-toc-wrapper">
<div class="btn-group w-100 mb-2" role="group" aria-label="rellinks">
<a href="../dispatching.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="11. Dispatching">Prev</a><a href="../dispatching.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="11. Dispatching">Up</a>
<a href="../metadata_routing.html" role="button" class="btn sk-btn-rellink py-1" sk-rellink-tooltip="1. Metadata Routing">Next</a>
</div>
<div class="alert alert-danger p-1 mb-2" role="alert">
<p class="text-center mb-0">
<strong>scikit-learn 1.5.dev0</strong><br/>
<a href="http://scikit-learn.org/dev/versions.html">Other versions</a>
</p>
</div>
<div class="alert alert-warning p-1 mb-2" role="alert">
<p class="text-center mb-0">
Please <a class="font-weight-bold" href="../about.html#citing-scikit-learn"><string>cite us</string></a> if you use the software.
</p>
</div>
<div class="sk-sidebar-toc">
<ul>
<li>
<a href="../user_guide.html" class="sk-toc-active">User Guide</a>
</li>
<ul>
<li>
<a href="../supervised_learning.html" class="">1. Supervised learning</a>
</li>
<li>
<a href="../unsupervised_learning.html" class="">2. Unsupervised learning</a>
</li>
<li>
<a href="../model_selection.html" class="">3. Model selection and evaluation</a>
</li>
<li>
<a href="../inspection.html" class="">4. Inspection</a>
</li>
<li>
<a href="../visualizations.html" class="">5. Visualizations</a>
</li>
<li>
<a href="../data_transforms.html" class="">6. Dataset transformations</a>
</li>
<li>
<a href="../datasets.html" class="">7. Dataset loading utilities</a>
</li>
<li>
<a href="../computing.html" class="">8. Computing with scikit-learn</a>
</li>
<li>
<a href="../model_persistence.html" class="">9. Model persistence</a>
</li>
<li>
<a href="../common_pitfalls.html" class="">10. Common pitfalls and recommended practices</a>
</li>
<li>
<a href="../dispatching.html" class="sk-toc-active">11. Dispatching</a>
<ul>
<li class="sk-toctree-l3">
<a href="">11.1. Array API support (experimental)</a>
</li>
</ul>
</li>
</ul>
</ul>
</div>
</div>
</div>
<div id="sk-page-content-wrapper">
<div class="sk-page-content container-fluid body px-md-3" role="main">
<section id="array-api-support-experimental">
<span id="array-api"></span><h1><span class="section-number">11.1. </span>Array API support (experimental)<a class="headerlink" href="#array-api-support-experimental" title="Link to this heading">¶</a></h1>
<p>The <a class="reference external" href="https://data-apis.org/array-api/latest/">Array API</a> specification defines
a standard API for all array manipulation libraries with a NumPy-like API.
Scikit-learn’s Array API support requires
<a class="reference external" href="https://github.com/data-apis/array-api-compat">array-api-compat</a> to be installed.</p>
<p>Some scikit-learn estimators that primarily rely on NumPy (as opposed to using
Cython) to implement the algorithmic logic of their <code class="docutils literal notranslate"><span class="pre">fit</span></code>, <code class="docutils literal notranslate"><span class="pre">predict</span></code> or
<code class="docutils literal notranslate"><span class="pre">transform</span></code> methods can be configured to accept any Array API compatible input
datastructures and automatically dispatch operations to the underlying namespace
instead of relying on NumPy.</p>
<p>At this stage, this support is <strong>considered experimental</strong> and must be enabled
explicitly as explained in the following.</p>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>Currently, only <code class="docutils literal notranslate"><span class="pre">cupy.array_api</span></code>, <code class="docutils literal notranslate"><span class="pre">numpy.array_api</span></code>, <code class="docutils literal notranslate"><span class="pre">cupy</span></code>, and <code class="docutils literal notranslate"><span class="pre">PyTorch</span></code>
are known to work with scikit-learn’s estimators.</p>
</div>
<section id="example-usage">
<h2><span class="section-number">11.1.1. </span>Example usage<a class="headerlink" href="#example-usage" title="Link to this heading">¶</a></h2>
<p>Here is an example code snippet to demonstrate how to use <a class="reference external" href="https://cupy.dev/">CuPy</a> to run
<a class="reference internal" href="generated/sklearn.discriminant_analysis.LinearDiscriminantAnalysis.html#sklearn.discriminant_analysis.LinearDiscriminantAnalysis" title="sklearn.discriminant_analysis.LinearDiscriminantAnalysis"><code class="xref py py-class docutils literal notranslate"><span class="pre">LinearDiscriminantAnalysis</span></code></a> on a GPU:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.datasets</span> <span class="kn">import</span> <span class="n">make_classification</span>
<span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn</span> <span class="kn">import</span> <span class="n">config_context</span>
<span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.discriminant_analysis</span> <span class="kn">import</span> <span class="n">LinearDiscriminantAnalysis</span>
<span class="gp">>>> </span><span class="kn">import</span> <span class="nn">cupy</span>
<span class="gp">>>> </span><span class="n">X_np</span><span class="p">,</span> <span class="n">y_np</span> <span class="o">=</span> <span class="n">make_classification</span><span class="p">(</span><span class="n">random_state</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_cu</span> <span class="o">=</span> <span class="n">cupy</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">X_np</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">y_cu</span> <span class="o">=</span> <span class="n">cupy</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">y_np</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_cu</span><span class="o">.</span><span class="n">device</span>
<span class="go"><CUDA Device 0></span>
<span class="gp">>>> </span><span class="k">with</span> <span class="n">config_context</span><span class="p">(</span><span class="n">array_api_dispatch</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
<span class="gp">... </span> <span class="n">lda</span> <span class="o">=</span> <span class="n">LinearDiscriminantAnalysis</span><span class="p">()</span>
<span class="gp">... </span> <span class="n">X_trans</span> <span class="o">=</span> <span class="n">lda</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">X_cu</span><span class="p">,</span> <span class="n">y_cu</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_trans</span><span class="o">.</span><span class="n">device</span>
<span class="go"><CUDA Device 0></span>
</pre></div>
</div>
<p>After the model is trained, fitted attributes that are arrays will also be
from the same Array API namespace as the training data. For example, if CuPy’s
Array API namespace was used for training, then fitted attributes will be on the
GPU. We provide a experimental <code class="docutils literal notranslate"><span class="pre">_estimator_with_converted_arrays</span></code> utility that
transfers an estimator attributes from Array API to a ndarray:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">sklearn.utils._array_api</span> <span class="kn">import</span> <span class="n">_estimator_with_converted_arrays</span>
<span class="gp">>>> </span><span class="n">cupy_to_ndarray</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">array</span> <span class="p">:</span> <span class="n">array</span><span class="o">.</span><span class="n">get</span><span class="p">()</span>
<span class="gp">>>> </span><span class="n">lda_np</span> <span class="o">=</span> <span class="n">_estimator_with_converted_arrays</span><span class="p">(</span><span class="n">lda</span><span class="p">,</span> <span class="n">cupy_to_ndarray</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">X_trans</span> <span class="o">=</span> <span class="n">lda_np</span><span class="o">.</span><span class="n">transform</span><span class="p">(</span><span class="n">X_np</span><span class="p">)</span>
<span class="gp">>>> </span><span class="nb">type</span><span class="p">(</span><span class="n">X_trans</span><span class="p">)</span>
<span class="go"><class 'numpy.ndarray'></span>
</pre></div>
</div>
<section id="pytorch-support">
<h3><span class="section-number">11.1.1.1. </span>PyTorch Support<a class="headerlink" href="#pytorch-support" title="Link to this heading">¶</a></h3>
<p>PyTorch Tensors are supported by setting <code class="docutils literal notranslate"><span class="pre">array_api_dispatch=True</span></code> and passing in
the tensors directly:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">torch</span>
<span class="gp">>>> </span><span class="n">X_torch</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">X_np</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="s2">"cuda"</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">y_torch</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">y_np</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="s2">"cuda"</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
<span class="gp">>>> </span><span class="k">with</span> <span class="n">config_context</span><span class="p">(</span><span class="n">array_api_dispatch</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
<span class="gp">... </span> <span class="n">lda</span> <span class="o">=</span> <span class="n">LinearDiscriminantAnalysis</span><span class="p">()</span>
<span class="gp">... </span> <span class="n">X_trans</span> <span class="o">=</span> <span class="n">lda</span><span class="o">.</span><span class="n">fit_transform</span><span class="p">(</span><span class="n">X_torch</span><span class="p">,</span> <span class="n">y_torch</span><span class="p">)</span>
<span class="gp">>>> </span><span class="nb">type</span><span class="p">(</span><span class="n">X_trans</span><span class="p">)</span>
<span class="go"><class 'torch.Tensor'></span>
<span class="gp">>>> </span><span class="n">X_trans</span><span class="o">.</span><span class="n">device</span><span class="o">.</span><span class="n">type</span>
<span class="go">'cuda'</span>
</pre></div>
</div>
</section>
</section>
<section id="support-for-array-api-compatible-inputs">
<span id="array-api-supported"></span><h2><span class="section-number">11.1.2. </span>Support for <code class="docutils literal notranslate"><span class="pre">Array</span> <span class="pre">API</span></code>-compatible inputs<a class="headerlink" href="#support-for-array-api-compatible-inputs" title="Link to this heading">¶</a></h2>
<p>Estimators and other tools in scikit-learn that support Array API compatible inputs.</p>
<section id="estimators">
<h3><span class="section-number">11.1.2.1. </span>Estimators<a class="headerlink" href="#estimators" title="Link to this heading">¶</a></h3>
<ul class="simple">
<li><p><a class="reference internal" href="generated/sklearn.decomposition.PCA.html#sklearn.decomposition.PCA" title="sklearn.decomposition.PCA"><code class="xref py py-class docutils literal notranslate"><span class="pre">decomposition.PCA</span></code></a> (with <code class="docutils literal notranslate"><span class="pre">svd_solver="full"</span></code>,
<code class="docutils literal notranslate"><span class="pre">svd_solver="randomized"</span></code> and <code class="docutils literal notranslate"><span class="pre">power_iteration_normalizer="QR"</span></code>)</p></li>
<li><p><a class="reference internal" href="generated/sklearn.discriminant_analysis.LinearDiscriminantAnalysis.html#sklearn.discriminant_analysis.LinearDiscriminantAnalysis" title="sklearn.discriminant_analysis.LinearDiscriminantAnalysis"><code class="xref py py-class docutils literal notranslate"><span class="pre">discriminant_analysis.LinearDiscriminantAnalysis</span></code></a> (with <code class="docutils literal notranslate"><span class="pre">solver="svd"</span></code>)</p></li>
<li><p><a class="reference internal" href="generated/sklearn.preprocessing.KernelCenterer.html#sklearn.preprocessing.KernelCenterer" title="sklearn.preprocessing.KernelCenterer"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.KernelCenterer</span></code></a></p></li>
<li><p><a class="reference internal" href="generated/sklearn.preprocessing.MaxAbsScaler.html#sklearn.preprocessing.MaxAbsScaler" title="sklearn.preprocessing.MaxAbsScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.MaxAbsScaler</span></code></a></p></li>
<li><p><a class="reference internal" href="generated/sklearn.preprocessing.MinMaxScaler.html#sklearn.preprocessing.MinMaxScaler" title="sklearn.preprocessing.MinMaxScaler"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.MinMaxScaler</span></code></a></p></li>
<li><p><a class="reference internal" href="generated/sklearn.preprocessing.Normalizer.html#sklearn.preprocessing.Normalizer" title="sklearn.preprocessing.Normalizer"><code class="xref py py-class docutils literal notranslate"><span class="pre">preprocessing.Normalizer</span></code></a></p></li>
</ul>
</section>
<section id="metrics">
<h3><span class="section-number">11.1.2.2. </span>Metrics<a class="headerlink" href="#metrics" title="Link to this heading">¶</a></h3>
<ul class="simple">
<li><p><a class="reference internal" href="generated/sklearn.metrics.accuracy_score.html#sklearn.metrics.accuracy_score" title="sklearn.metrics.accuracy_score"><code class="xref py py-func docutils literal notranslate"><span class="pre">sklearn.metrics.accuracy_score</span></code></a></p></li>
<li><p><a class="reference internal" href="generated/sklearn.metrics.zero_one_loss.html#sklearn.metrics.zero_one_loss" title="sklearn.metrics.zero_one_loss"><code class="xref py py-func docutils literal notranslate"><span class="pre">sklearn.metrics.zero_one_loss</span></code></a></p></li>
</ul>
</section>
<section id="tools">
<h3><span class="section-number">11.1.2.3. </span>Tools<a class="headerlink" href="#tools" title="Link to this heading">¶</a></h3>
<ul class="simple">
<li><p><a class="reference internal" href="generated/sklearn.model_selection.train_test_split.html#sklearn.model_selection.train_test_split" title="sklearn.model_selection.train_test_split"><code class="xref py py-func docutils literal notranslate"><span class="pre">model_selection.train_test_split</span></code></a></p></li>
</ul>
<p>Coverage is expected to grow over time. Please follow the dedicated <a class="reference external" href="https://github.com/scikit-learn/scikit-learn/issues/22352">meta-issue on GitHub</a> to track progress.</p>
</section>
</section>
<section id="common-estimator-checks">
<h2><span class="section-number">11.1.3. </span>Common estimator checks<a class="headerlink" href="#common-estimator-checks" title="Link to this heading">¶</a></h2>
<p>Add the <code class="docutils literal notranslate"><span class="pre">array_api_support</span></code> tag to an estimator’s set of tags to indicate that
it supports the Array API. This will enable dedicated checks as part of the
common tests to verify that the estimators result’s are the same when using
vanilla NumPy and Array API inputs.</p>
<p>To run these checks you need to install
<a class="reference external" href="https://github.com/data-apis/array-api-compat">array_api_compat</a> in your
test environment. To run the full set of checks you need to install both
<a class="reference external" href="https://pytorch.org/">PyTorch</a> and <a class="reference external" href="https://cupy.dev/">CuPy</a> and have
a GPU. Checks that can not be executed or have missing dependencies will be
automatically skipped. Therefore it’s important to run the tests with the
<code class="docutils literal notranslate"><span class="pre">-v</span></code> flag to see which checks are skipped:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><style type="text/css">
span.prompt1:before {
content: "$ ";
}
</style><span class="prompt1">pip<span class="w"> </span>install<span class="w"> </span>array-api-compat<span class="w"> </span><span class="c1"># and other libraries as needed</span></span>
<span class="prompt1">pytest<span class="w"> </span>-k<span class="w"> </span><span class="s2">"array_api"</span><span class="w"> </span>-v</span>
</pre></div></div><section id="note-on-mps-device-support">
<h3><span class="section-number">11.1.3.1. </span>Note on MPS device support<a class="headerlink" href="#note-on-mps-device-support" title="Link to this heading">¶</a></h3>
<p>On macOS, PyTorch can use the Metal Performance Shaders (MPS) to access
hardware accelerators (e.g. the internal GPU component of the M1 or M2 chips).
However, the MPS device support for PyTorch is incomplete at the time of
writing. See the following github issue for more details:</p>
<ul class="simple">
<li><p><a class="reference external" href="https://github.com/pytorch/pytorch/issues/77764">https://github.com/pytorch/pytorch/issues/77764</a></p></li>
</ul>
<p>To enable the MPS support in PyTorch, set the environment variable
<code class="docutils literal notranslate"><span class="pre">PYTORCH_ENABLE_MPS_FALLBACK=1</span></code> before running the tests:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span class="prompt1"><span class="nv">PYTORCH_ENABLE_MPS_FALLBACK</span><span class="o">=</span><span class="m">1</span><span class="w"> </span>pytest<span class="w"> </span>-k<span class="w"> </span><span class="s2">"array_api"</span><span class="w"> </span>-v</span>
</pre></div></div><p>At the time of writing all scikit-learn tests should pass, however, the
computational speed is not necessarily better than with the CPU device.</p>
</section>
</section>
</section>
</div>
<div class="container">
<footer class="sk-content-footer">
© 2007 - 2023, scikit-learn developers (BSD License).
<a href="../_sources/modules/array_api.rst.txt" rel="nofollow">Show this page source</a>
</footer>
</div>
</div>
</div>
<script src="../_static/js/vendor/bootstrap.min.js"></script>
<script>
window.ga=window.ga||function(){(ga.q=ga.q||[]).push(arguments)};ga.l=+new Date;
ga('create', 'UA-22606712-2', 'auto');
ga('set', 'anonymizeIp', true);
ga('send', 'pageview');
</script>
<script async src='https://www.google-analytics.com/analytics.js'></script>
<script defer data-domain="scikit-learn.org" src="https://views.scientific-python.org/js/script.js">
</script>
<script src="../_static/clipboard.min.js"></script>
<script src="../_static/copybutton.js"></script>
<script>
$(document).ready(function() {
/* Add a [>>>] button on the top-right corner of code samples to hide
* the >>> and ... prompts and the output and thus make the code
* copyable. */
var div = $('.highlight-python .highlight,' +
'.highlight-python3 .highlight,' +
'.highlight-pycon .highlight,' +
'.highlight-default .highlight')
var pre = div.find('pre');
// get the styles from the current theme
pre.parent().parent().css('position', 'relative');
// create and add the button to all the code blocks that contain >>>
div.each(function(index) {
var jthis = $(this);
// tracebacks (.gt) contain bare text elements that need to be
// wrapped in a span to work with .nextUntil() (see later)
jthis.find('pre:has(.gt)').contents().filter(function() {
return ((this.nodeType == 3) && (this.data.trim().length > 0));
}).wrap('<span>');
});
/*** Add permalink buttons next to glossary terms ***/
$('dl.glossary > dt[id]').append(function() {
return ('<a class="headerlink" href="#' +
this.getAttribute('id') +
'" title="Permalink to this term">¶</a>');
});
});
</script>
<script id="MathJax-script" async src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-chtml.js"></script>
<script src="https://scikit-learn.org/versionwarning.js"></script>
</body>
</html>